4. Electronic Theses and Dissertations (ETDs) - Faculties submissions

Permanent URI for this community

https://hdl.handle.net/10539/37773

Browse

Now showing 1 - 20 of 207

A Beginner’s Guide to Puzzle Design: Creating an applied guide for effective puzzle design in videogames
(University of the Witwatersrand, Johannesburg, 2023-06) Prinz, Erik; Flusk, Timothy; Reid, Kieran
While a vast collection of information exploring effective puzzle design exists, it is riddled with conflicting opinions and inconsistent formats, making it arduous to engage with. This research aims to curate this collection, organizing its knowledge into an applied guide for puzzle design in videogames. This will be accomplished through an amalgamation of design principles offered by selected works of oundational literature, and the insights contained within the design philosophies of three industry professionals (Jonathan Blow, Kim Swift, and Arvi ‘Hempuli’ Teikari). The synthesis of these two families of information will be aided by the simultaneous development of a puzzle platforming videogame. Through a process of iteration and playtesting, this game will be used to assess the value and accuracy of the developing guide. With careful consideration of useful structures present in the foundational literature, the plethora of existing information can be reformatted to be user-friendly and appliable by novel puzzle designers.
A comparative analysis of transformation between local and global media and advertising agencies in South Africa
(University of the Witswatersrand, Johannesburg, 2023) Ndinguri, Kevin
Poverty, unemployment and inequality among black people have been identified as key consequences of apartheid. To redress these, the Broad-Based Black Economic Empowerment Act was enacted. In the Media and Advertising industry in South Africa, the B-BBEE Act is translated through the Media and Communications Charter (MAC) which describes five elements, namely; management inclusion, skills development, responsible advertising, preferential procurement and socio-economic development. Most of the media industry studies on the compliance to the MAC charter focus on descriptions of how companies have applied the B-BBEE Act. This study’s specific objectives were to describe the nature of transformation in local media and advertising agencies, describe the nature of transformation in global media and advertising agencies and describe the effect of B-BBEE codes on local and global media and advertising agencies in South Africa. The study used a qualitative approach and an exploratory-descriptive design. The data collection tool was semi-structured interviews and data was collected through in-depth interviews. Data were analysed using thematic content analysis and presented in narrative format.The study found four themes that compare the transformation of local and global media and advertising agencies. The first theme is that transformation has taken place in local media agencies, the second is that few transformation initiatives have taken place in global media companies. The third theme is that B-BBEE codes have enabled local media companies to secure clients and the fourth theme is that B-BBEE codes have obligated global media agencies to transform the way they manage their businesses. In conclusion, global media and advertising agencies have been affected by transformation through increased diversity and change in the way they managed their businesses. Recommendations for future research should use a quantitative approach to enable the generalisation of findings
A Comparative Study of the Efficiencies of Vertical Bracing Practices
(University of the Witwatersrand, Johannesburg, 2023-11) Saunders, Shawn Wayne Valintino; Elvin, Alex
The efficiencies of cross sections and configurations applied to vertical bracing are investigated by evaluating reference configurations (RCs), composed of cross braced circular hollow section (CHS) members, against comparative configurations (CCs), consisting of cross-braced Angle members, and single-CHS members. The metrics used to evaluate efficiencies were mass, raw materials costs, and fabrication and erection costs. CCs were found to be more efficient than RCs for most analysed cases, metric and configuration dependent. The following results were found: i. Mass metric a. Crossed-Angle more efficient in 79% of analysed cases. b. Single CHS more efficient in 87% of analysed cases. ii. Raw materials costs metric a. Crossed-Angle more efficient in 92% of analysed cases b. Single CHS more efficient in 88% of analysed cases iii. Fabrication and erection costs metric a. Crossed-Angle more efficient in 90.4% of analysed cases b. Single CHS more efficient in 88.5% of analysed cases Inversions of the efficiency parameter findings, with RCs more efficient than CCs, were observed when: i. RC CHS member slenderness ratios were less than 80-90. ii. CC design loads were greater than 225 kN, 1200 kN and 1500 kN for mass, raw materials and total cost efficiency metrics, respectively
A digital fraud strategy framework for banks and insurance companies in South Africa
(University of the Witwatersrand, Johannesburg, 2022-08) Mahlangu, Cecilia Nonhlanhla Pretty; Oro, Oro Ufuo
South African banks and insurance companies have been focused on transforming and accelerating their digital strategies. There has been much focus for most organisations on providing products or services across various digital platforms. As a result, fraudsters are also finding new ways to commit fraud on the same digital platforms. Consequently, digital fraud numbers have increased, and banks and insurance companies should ensure that they implement suitable measures that will help them predict, prevent, detect, monitor, and respond to digital fraud in real time while protecting their customers' financial goals and improving customer experience. Existing studies have revealed components of combating digital fraud for the South African financial service providers, but no study on a holistic digital fraud strategy framework has been conducted. The aim of this research is therefore to evaluate a digital fraud strategy framework for South African banks and insurance companies, and to obtain a better understanding of how to combat digital fraud for South African banks and insurance companies. A mixed-methods approach was used in this study, in which participants in the banking and insurance industry were surveyed, and digital fraud subject matter experts were interviewed. Data analysis revealed that organisational structure, education and awareness, policies, big data, and analytics all have an impact on reducing digital fraud for South African banking and insurance institutions. The study also found that government legislation does not affect the reduction of digital fraud. South African banking and insurance institutions therefore need to implement appropriate measures that will help combat digital fraud. Keywords:
A Digital Platform Car Wash Business
(University of the Witwatersrand, Johannesburg, 2023) Ndlovu, Cebo
The Car Wash Digital Platform (App) is a new business venture that addresses the problem of customers needing a convenient car wash service. This study was aimed at evaluating the viability of this new business venture. A customer discovery was conducted using an online survey to test if potential customers would be interested in such a service. The customer discovery was only made on the end user customer who would request the car wash service. The customer discovery for the service providers was not completed due to time limitations. This project's competitive advantages include the convenience it will provide to its clients in addition to fuel and mileage saved by not having to drive to a particular car wash facility. The time saved from driving to and from the car wash service provider is another competitive advantage of this business venture. The New Business Venture adopted the Lean Business Canvas as its business strategy model. The data from the online survey contributed to mapping the Lean Business Canvas. The marketing strategy was developed by using the marketing mix of customer segmentation, the 4 Ps of marketing, and Porter’s five forces model. The financial analyses indicated that the new business venture could break even in its first year of operation and generate an estimated Net Profit Margin of 40% by the fourth year
A framework for Mobile Application integrated with Biometric Authentication to Improve Youth participation in Elections
(University of the Witswatersrand, Johannesburg, 2023) Botsime, Kealeboga Obed; Pellissier, Prof René
The purpose of this research study is to apply the technology acceptance framework to assess the prospect of using mobile application integrated with biometric authentication that might improve the South African’s youth participation in elections. The main objectives of this research are to measure the extent to which youth voters have trust in internet and the ability for the national government to conduct online elections, evaluate the youth voters ICT accessibility and skills and to determine the attitudes, perceptions as well as intention to participate in online elections using framework for mobile application. A quantitative research approach was adopted for this research. An onlinequestionnaire on formplus platform is used to collect data from Gauteng youth (18-35 years); 209 participants had valid responses. A conceptual framework is proposed using constructs and theories from the Technology Acceptance Model (TAM). The constructs were adopted from the Technology Acceptance Model (TAM), and it was also extended with trust constructs. The findings of this research are youth respondents have indicated that they agree (eighty-four percent) and have intention to use the mobile application to participate in online voting if it becomes available and they have the willingness to learn how to use it. The intention of youth to use mobile phones for online voting as an important contributing factor to improve youth participation in elections.
A fully-decentralised general-sum approach for multi-agent reinforcement learning using minimal modelling
(University of the Witwatersrand, Johannesburg, 2023-08) Kruger, Marcel Matthew Anthony; Rosman, Benjamin; James, Steven; Shipton, Jarrod
Multi-agent reinforcement learning is a prominent area of research in machine learning, extending reinforcement learning to scenarios where multiple agents concurrently learn and interact within the same environment. Most existing methods rely on centralisation during training, while others employ agent modelling. In contrast, we propose a novel method that adapts the role of entropy to assist in fully-decentralised training without explicitly modelling other agents using additional information to which most centralised methods assume access. We augment entropy to encourage more deterministic agents, and instead, we let the non-stationarity inherent in MARL serve as a mode for exploration. We empirically evaluate the performance of our method across five distinct environments, each representing unique challenges. Our assessment encompasses both cooperative and competitive cases. Our findings indicate that the approach of penalising entropy, rather than rewarding it, enables agents to perform at least as well as the prevailing standard of entropy maximisation. Moreover, our alternative approach achieves several of the original objectives of entropy regularisation in reinforcement learning, such as increased sample efficiency and potentially better final rewards. Whilst entropy has a significant role, our results in the competitive case indicate that position bias is still a considerable challenge.
A legal approach to whether ai generated content should be protected under copyright
(University of the Witswatersrand, Johannesburg, 2023-10) Stransky, Caitlin
Currently, there is no clear answer for whether AI-generated content should be protected under copyright law in South Africa and if so, who is the author and who owns the copyright. As AI is growing more advanced and widely used, the potential for confusion grows and thus clarity on the law’s position is important. In this paper, I will examine how AI generates works and compare this to the purpose of copyright law. I will then determine whether granting AI- generated works copyright protection aligns with the purpose of copyright law. Further, I will conduct an analysis on how different countries have dealt with the question of whether AI-generated content should be protected under copyright. This will help pinpoint the factors to consider when answering this question such as the requirement of human authorship, legal or juristic person ownership of the copyright and whether the works are copyrightable. It is important to consider how South Africa’s legal system should approach issues surrounding AI-generated content and copyright. In examining the South African legal position on this question, I will conduct an analysis on the approaches taken by the US, the UK, Germany, Australia, China, and South Africa in relation to AI-generated works and copyright. This range of countries will allow for a greater understanding of the issues, complexities, and factors to consider while answering this question
A needs assessment of outsourcing information technology and business services to rural areas in South Africa
(University of the Witwatersrand, Johannesburg, 2020) Ramagondo, Vhumbani; Louwrens, Jessica
The purpose of this study was to investigate the perception of South African organisations on the rural outsourcing of their Business Processing and Information Technology services. This study is based on a questionnaire survey conducted on South African organisations. The goal of this survey is to determine the perceptions of people concerning the attractiveness of potentially outsourcing their services to rural areas. Results of the study show a negative perception towards the attractiveness of rural outsourcing. The results also highlighted the factors that are contributing to this negative perception of rural outsourcing. The findings of this study have several implications for the viability of a rural outsourcing market in South Africa. The results also provide both potential rural outsourcing suppliers and clients with an assessment of the potential of rural outsourcing business processing outsourcing (BPO) and information technology outsourcing (ITO). The study also makes a valuable contribution to the outsourcing industry by identifying the business process and information technology services that are most suitable for rural outsourcing
Adoption and use of internet of things and the implications for adaptive regulation
(University of the Witwatersrand, Johannesburg, 2024-03) Jarvis, Boipelo; Ntlatlapa, Ntsibane
Internet of Things (IoT) is evolving, developing and finding use in many industries where it is mainly used for automating, controlling, tracking and monitoring of different assets and processes, and also to digitalise and optimise business processes. One of IoT’s main characteristics is the interconnection of physical and virtual objects, the involvement of various stakeholders and the vast amount of data that is collected, communicated, stored and analysed in its ecosystem. IoT is projected to continue on its tremendous growth path for years to come, and to also permeate many more industries. However, IoT has inherent challenges of security and privacy due to its characteristics and therefore requires relevant regulation so as to address the challenges related to it and enable its continued growth, adoption and use. The study explored the adoption and use of IoT in South Africa, looked into security and privacy challenges for IoT and ways to address them, ways in which current regulatory approaches are affecting IoT and how regulation that is relevant to IoT can be developed. The researcher followed a qualitative research approach, collected data from participants through in-depth interviews and employed thematic analysis to discover themes from data that was collected. The study’s findings are categorised according to three themes that emerged from the literature review namely: adoption and use of IoT, security and privacy challenges for IoT and ways to develop adaptive regulation for IoT. The Socio Technical Systems (STS) framework was a theoretical lens that was used to analyse data by mapping the study’s findings against STS components to explore the social and technical aspects of IoT. To define and understand the relationship between the social and technical subsystems of STS, an interaction between the elements of these two subsystems namely technology, tasks, structure and people was done. Under the technical subsystem, the findings brought to light the challenge that interoperability, security and privacy has on the adoption and use of IoT and the activities that enable the adoption and use of IoT. The findings under the social subsystem highlighted regulatory measures that are required to enable the adoption and use of IoT, regulatory ways to address the challenges of security and privacy as well as the need for IoT stakeholders to work collaboratively to encourage the growth, adoption and use of IoT and to address challenges related to it. The interaction of the STS elements identified collaborations and collaborative mechanisms as ways to address the challenges of IoT and develop regulation that is adaptive to its development. Based on the researcher’s analysis IoT requires a collaborative approach to address the challenges that its development, adoption and use are confronted with and to also develop regulation that is relevant and encouraging of its adoption and use.
Adoption of embedded subscriber identity module (ESIM) technology services by South African smartphone users for international roaming
(University of the Witwatersrand, Johannesburg, 2023) Singh, Sharon
The objective of this study was to investigate the factors that influence the intention to adopt ESIM (embedded SIM) technology for international roaming services by South African smartphone users. Currently, the Subscriber Identity Module (SIM) is a physical hardware with an integrated chipset that is inserted into mobile handsets. Physical SIM cards allow a single profile to be provisioned for users. The next evolution in SIM card technology, ESIM, have the capability to support multiple user profiles. The development of services and new business models enabled through ESIM technology is dependent on the willingness of users to adopt these services. A theoretical model was developed using TAM as a basis with the additional constructs of trust and risk to test the intention to adopt international roaming services. A cross-sectional, quantitative study using an online survey instrument was employed. A sample size of 142 was collected. The sampling method used was a non-probability, convenience sampling technique and Confirmatory Factor Analysis (CFA) was performed to examine the reliability and validity of the measures and the main conceptual model was tested using structural equation modelling (SEM). Perceived ease of use and perceived usefulness were identified as the most significant factors that influence the intention to adopt ESIM roaming services. Perceived usefulness was identified as significant mediating variable between perceived ease of use and intention to adopt. Perceived risk presented as statistically significant, direct influences on intention to adopt. The trust factor had an insignificant influence of the intention to adopt. The study’s findings form a basis for future research on ESIM enabled services. These findings are relevant to mobile operators, technology platform owners and developers of services for ESIM enabled devices (smartphones, wearables and internet of things devices) with the intent to understand factors that drive users’ intention to adopt service
The adoption of intelligent robot automation by auditors within South Africa
(2021) Naidoo, Evaashan
To automate, or not to automate, that is the question? The use of disruptive technologies has changed, and in many instances, has replaced traditional business operations and models. The auditing profession has been burdened with high costs and reputational damage as a result of false results due to a high dependency on manual tasks that are susceptible to errors or manipulation. The use of robots to help minimise these errors, achieve efficiencies and reduce costs, provides a convincing theoretical case to automate. The objective of this research is to apply the Unified Theory of Acceptance and Use of Technology (UTAUT) model to understand the views of audit professionals determine the factors that would influence organisations to use robot technology for performing audits in South Africa. The study adopted a quantitative approach, where a survey questionnaire consisting of 39 questions, aligned with the UTAUT model, was used to gather data. The sample comprised 59 auditors and 26 non-auditors who were involved in audits performed in South Africa. The non-auditors comprised of a Chief Information Officer, Finance Managers, Heads of Product, and various Senior Risk Management Professionals. A purposive sampling technique was used to select individuals who had an understanding of the capabilities of robots, and the potential use of robots to support business operations. The results of the study highlighted that performance expectancy and facilitating conditions were the key factors that influenced the adoption of robots in audits. Auditors and clients were more likely to use robots if they felt it would improve their performance. The respondents were more likely to use robots if it enabled the quicker completion of tasks, reduced errors, increased the number of tasks that could be completed and delivered better value than manual processes. The study also highlighted that sufficient investment and management support would be provided to ensure that an enabling environment is established for using robots to perform audit tasks. This would include acquiring management support as well as finding hardware, software, and skills to support the robotics development initiatives. The study provides more insight on how leaders and management in the auditing field can use insights attained through this study to successfully influence and support the use of robots to perform audits. Furthermore, the study contributes to an increased used of robotics in auditing which could improve the quality and efficiency of the auditor’s work. The study contributes towards the body of knowledge in the accounting and robot fields of study
Adoption of locally developed Metal Additive Manufacturing Technology: Evaluation of South Africa’s manufacturing industry’s readiness
(University of the Witwatersrand, Johannesburg, 2021) Sono, Oscar James Hendrick; Wotela, Kambidima; George, James
Presently, Metal Additive Manufacturing (MAM) has progressed to the stage where the manufacture of components for industries such aerospace, energy, automotive and medical industries, are possible to a high degree of reliability. The South African government has invested substantially on the technology’s advancement through various government entities, publicly funding the research and development programmes and infrastructure developments. Yet the adoption of the technology remains slow, contrast to the USA and other European countries. Therefore, this thesis sought to explore the readiness of South Africa’s manufacturing industry to adopt locally developed MAM technologies. This was undertaken by employing a quantitative research approach, in which a cross- sectional design survey was used to conduct the interrogation, guided by innovation diffusion theories, particularly DOI. The study applies descriptive analysis to determine the current state of MAM adoption and the level of awareness about the technology within the industry. Backward multiple regression was used to determine if cost, organizational capability, and potential job loss have an impact on the adoption of the technology within the industry testing the three hypotheses. The results indicated that the state of MAM adoption within the manufacturing industry was low, with only three machines owned/leased, and potential adopters that are likely to uptake the technology, believing it can bring value-add to their organisations. These potential adopters are familiar with awareness programmes driving MAM campaigns. On the other hand, they also indicated reasons they believed would lead to non-uptake of the technology, the non-adopters did too, with cost of material being the leading constraint. Furthermore, results of tested hypotheses inferred a causal relationship between the adoption of MAM technology and the variables, and the impact was significant on its uptake or non-uptake. However, it was important to note that the sample size was low to take a conclusive stance, hence the researcher recommends a continuation of the study
Alternative Tenure Options as a Stepping Stone for Urban Land Reform: The Case of Kwa Mai Mai Muthi Market
(University of the Witwatersrand, Johannesburg, 2024-02) Mnguni, Thobeka; Maina, Miriam M.
This research uses qualitative methods to investigate the case of the Kwa Mai Mai muthi market located in the central business district of Johannesburg where over the decades informal traders have converted their trading units into residential homes. This case is used to respond to the research question, ‘To what extent is taking administrative action to improve land tenure by offering communities rights to occupy urban land feasible and necessary for advancing the urban land reform agenda?.’ The research acknowledges that popular approaches to improving tenure by converting insecure tenure rights to statutory ownership through the Deeds office has had limited success and resulted in efficiencies due to the high costs and the resources intensive nature of this approach. Instead, the research recommends a necessary transition to improving tenure in innovative, context-specific ways that work to build on existing settlements through taking low-cost and low LOE administrative action.
An assessment of the intention to adopt blockchain technology (BCT) in SMEs: A South African Analysis
(University of the Witwatersrand, Johannesburg, 2023) Anderson, Brain
This study examines the factors influencing the intention to adopt blockchain technology (BCT) among small and medium-sized enterprises (SMEs) in South Africa. Despite the potential benefits of BCT, such as increased transparency, security, and efficiency, its adoption among businesses, especially SMEs in South Africa, is still relatively low. To address this gap, the study proposes a conceptual framework that extends the Technology Acceptance Model (TAM) by including both internal and external variables. The study hypothesizes that these variables influence SMEs' perceived usefulness and perceived ease of use of BCT, which, in turn, influences their intention to adopt the technology. The internal variables in the framework include innovativeness, self-efficacy, and perceived system complexity, while the external variables include strategic orientation and social influence. To test the hypotheses, the study collects data from a survey of SMEs in South Africa, using a structured questionnaire. The survey includes questions on the SMEs' familiarity with BCT, their perceptions of its benefits and drawbacks, and their intentions to adopt the technology. The data is analysed using structural equation modelling to test the relationships between the variables. The findings of the study suggest that SMEs' intentions to adopt BCT are influenced by their perceptions of the technology's usefulness and ease of use, as well as their innovativeness, self-efficacy, and perceptions of system complexity. The study also finds that external factors, such as strategic orientation and social influence, play a significant role in influencing SMEs' intentions to adopt BCT. Overall, the study contributes to the literature on BCT adoption by extending the TAM with internal and external variables that are specific to the South African context. The study's findings have practical implications for SMEs in South Africa that are considering adopting BCT, as it can help them understand the factors that influence their intentions to adopt the technology. The study also provides recommendations for researchers who are interested in promoting the adoption of BCT among SMEs in South Africa.
An integrated approach for detecting and monitoring the Campuloclinium macrocephalum (Less) DC using the MaxEnt and machine learning models in the Cradle Nature Reserve, South Africa
(University of the Witwatersrand, Johannesburg, 2024) Makobe, Benjamin; Mhangara, Paidamoyo
The invasion of ecosystems by invasive plants is considered as one of the major human- induced global environmental change. The uncontrolled expansion of invasive alien plants is gaining international attention, and remote sensing technology is adopted to accurately detect and monitor the spread of invasive plants locally and globally. The Greater Cradle nature reserve is a world heritage site and intense research site for archaeology and paleontology.It was accorded the world status by the United Nations Educational, Scientific and Cultural Organizations (UNESCO) in 1991 due to its variety of biodiversity present and carries information of significance about the evolution of mankind. The invasion of Campuloclinium macrocephalum (pompom weed) at the Cradle nature reserve is downgrading the world status accorded to the site, lowers the grazing capacity for game animals and replaces the native vegetation. This research study explored the capability of Sentinel-2A multispectral imagery in mapping the spatial distribution of pompom weed at the nature reserve between 2019 and 2024. The non-parametric classification models, support vector machine (SVM) and random forests (RF) were evaluated to accurately detect, and discriminate pompom weed against the co-existing land cover types. Additionally, the species distribution modelling MaxEnt Entropy was incorporated to model spatial distribution and pompom weed habitat suitability. The findings indicates that SVM yielded 44% and 50.7% spatial coverage of pompom weed at the nature reserve in 2019 and 2024, respectively. Whereas, the RF model indicates that the spatial coverage of pompom weed was 31.1% and 39.3% in 2019 and 2024, respectively. The MaxEnt model identified both soil and rainfall as the most important environmental factors in fostering the aggressive proliferation of pompom weed at nature reserves. The MaxEnt predictive model obtained an area under curve score of 0.94, indicating outstanding prediction model performance. SVM and RF models had classification accuracy above 75%, indicating that they could distinguish pompom weeds from existing land cover types. The preliminary results of this study call for attention in using predictive models in predicting current and future spatial distribution of invasive weeds, for effective eradication control and environmental management.
An investigation of the relationship between ICT infrastructure and economic growth of emerging market
(University of the Witwatersrand, Johannesburg, 2023-02) Jiang, Jun Wen; Fasanya, Ismail
The study examines the link between Information and Communication Technology, institutional quality, and economic growth in emerging markets over the period of 2000 to 2019, using the system Generalized Method of Moments. The connection between economic growth and technology lies on the framework of exogenous growth model. The following findings are discernible from the study. First, a substantial positive relationship exists between internet usage and economic growth, while a negative association between economic growth and fixed telephone users is evident. Second, a positive association between growth and innovation exist in emerging markets, whilst institutions reveal a negative association. These findings have a significant policy implication for policymakers to monitor innovation factors rather than institutional quality to bypass the digital divide. Consequently, policymakers should pay attention to the benefits of Information and Communication technology usage by means of reducing entries cost whilst improving network facilities transfers
Analysing the player's involvement in video game character animation
(University of the Witwatersrand, Johannesburg, 2021) Ponde, Rugare; Whitcher, Ray
This dissertation investigates the impact of animation on player interest in an existing video game character. There is high demand and expectation regarding character animation quality, yet understanding and achieving this standard is complex and challenging. The quality of a character’s animation is often derived from its direct appeal to the audience and how its representation informs their impression. However, there is a gap in the literature where there is little on how style representation plays a role in character perception and identification in video games. A practice-led approach was used to understand the relationship between player involvement and the player-character's appeal. The process involved the creation of an animation reel to demonstrate how a popular video game character's acceptance can change based on a player's perception of style in the character's animation. I used Link from the video game Super Smash Brothers Ultimate (Nintendo, 2018). The reel was presented to adults between 18 and 35 to review the animations. Using the process of creative exegesis, the theories and concepts about character appeal, animation design, and player involvement were combined to analyse and critique the contributing factors that inform the perception of the creative work. The results from this study indicate that a change in movement style impacted the perception of Link and the participant's demonstrated interest to play him. This study confirms that style representations are an important design consideration to improve a character’s appeal. This topic may benefit the art and technique of character design and how to improve on it.
Applying Machine Learning to Model South Africa’s Equity Market Index Price Performance
(University of the Witwatersrand, Johannesburg, 2023-07) Nokeri, Tshepo Chris; Mulaudzi, Rudzani; Ajoodha, Ritesh
Policymakers typically use statistical multivariate forecasting models to forecast the reaction of stock market returns to changing economic activities. However, these models frequently result in subpar performance due to inflexibility and incompetence in modeling non-linear relationships. Emerging research suggests that machine learning models can better handle data from non-linear dynamic systems and yield outstanding model performance. This research compared the performance of machine learning models to the performance of the benchmark model (the vector autoregressive model) when forecasting the reaction of stock market returns to changing economic activities in South Africa. The vector autoregressive model was used to forecast the reaction of stock market returns. It achieved a mean absolute percentage error (MAPE) value of 0.0084. Machine learning models were used to forecast the reaction of stock market returns. The lowest MAPE value was 0.0051. The machine learning model trained on low economic data dimensions performed 65% better than the benchmark model. Machine learning models also identified key economic activities when forecasting the reaction of stock market returns. Most research focused on whole features, few models for comparison, and barely focused on how different feature subsets and reduced dimensionality change model performance, a limitation this research addresses when considering the number of experiments. This research considered various experiments, i.e., different feature subsets and data dimensions, to determine whether machine learning models perform better than the benchmark model when forecasting the reaction of stock market returns to changing economic activities in South Africa.
Assessing State Capacity in South African Industrial Policy Design and Implementation in a Changing Global Landscape
(University of the Witwatersrand, Johannesburg, 2022) Diale, Rapula Comfort; Nkunzi, Sibulele
South Africa has sought to transform and reindustrialise its economy. Through its industrial policy, it has intervened in several manufacturing sub-sectors. Despite the numerous state interventions, the South African manufacturing sector continues to deteriorate. The shorting comings of industrial policy have raised fresh questions of whether the Department of Trade Industry and Competition (DTIC) and the state, including its private sector partners, have sufficient capacity to design and implement appropriate industrial policy. This paper aims to assess state capacity in South African industrial policy design and implementation. In so doing, it analyses developmental economic literature, including the Weberian state, the developmental state, the human capacity theory and the bureaucratic rationale theory. The evidence for this research was collected through interviews with key stakeholders in the industrial policy, including thought leaders, policy researchers, policymakers, and academics. The paper finds that to achieve successful policy intervention; the country needs to develop supportive institutions and systems in industrial policy across government departments and different spheres of government. It further argues that the state outsourcing policy research capacity, if used correctly, can be developmental; however, the incorrect usage of outsourced capacity is detrimental to state capacity

Browse

Browsing 4. Electronic Theses and Dissertations (ETDs) - Faculties submissions by SDG "SDG-9: Industry, innovation and infrastructure"

Results Per Page

Sort Options